Relationship between High-frequency Radiation and Asperity Ruptures, Revealed by Hybrid Back-projection with a Non-planar Fault Model

Abstract

High-frequency seismic waves are generated by abrupt changes of rupture velocity and slip-rate during an earthquake. Therefore, analysis of high-frequency waves is crucial to understanding the dynamic rupture process. Here, we developed a hybrid back-projection method that considers variations in focal mechanisms by introducing a non-planar fault model that reflects the subducting slab geometry. We applied it to teleseismic P-waveforms of the Mw 8.8 2010 Chile earthquake to estimate the spatiotemporal distribution of high-frequency (0.5–2.0 Hz) radiation. By comparing the result with the coseismic slip distribution obtained by waveform inversion, we found that strong high-frequency radiation can precede and may trigger a large asperity rupture. Moreover, in between the large slip events, high-frequency radiation of intermediate strength was concentrated along the rupture front. This distribution suggests that by bridging the two large slips, this intermediate-strength high-frequency radiation might play a key role in the interaction of the large slip events.

Rupture Process During the 2015 Illapel, Chile Earthquake: Zigzag-Along-Dip Rupture Episodes

“…The HBP method improves upon the backprojection (BP) method KRÜ GER and OHRNBERGER 2005) in two ways: it can mitigate the systematic delay of projected images of high-frequency sources distorted by the depth phases (pP and sP phases), and by using globally observed waveforms it can produce higher-resolution images than array-based BP methods (e.g., WALKER et al 2005;OKUWAKI et al 2014;FAN and SHEARER 2015). The HBP method can be useful for capturing the rupture front velocity, rupture direction, and rupture extents as it does not require values of these characteristics to be assumed before processing, and the information of rupture behaviors from the HBP method can be used as constraint conditions for the kinematic waveform inversion .…”

“…To construct a seismic source model, we estimated the spatiotemporal distribution of slip and high-frequency (0.3-2.0 Hz) sources by using the kinematic waveform inversion method that takes into account the uncertainty in Green's function (YAGI and FUKA-HATA 2011) together with the HBP method (YAGI et al 2012;OKUWAKI et al 2014) to track the spatiotemporal distribution of high-frequency sources, respectively. Our integrated approach, using a wide range of frequency contents in the P-waveforms, is essential for tracking the rupture propagation history in detail because the high-frequency waves (*1 Hz) are generated by abrupt changes in the rupture velocity and/or slip-rate (e.g., MADARIAGA 1977;BERNARD and MADARIAGA 1984;SPUDICH and FRAZER 1984), and the high-frequency signal can be an index Black dots correspond to the 1-week-aftershock epicenters matching criteria of M C 3 and shallower than 50 km depth, determined by the CSN.…”

Detailed rupture imaging of the 25 April 2015 Nepal earthquake using teleseismic P waves

“…However, in principle, higher resolution can be obtained using globally distributed stations [e.g., Walker et al, 2005;Yagi et al, 2012;Okuwaki et al, 2014] and this is the approach we adopt here. We analyze both a high-frequency band (0.2 to 3 Hz), similar to that used most often in prior backprojection studies, and a low-frequency band (0.05 to 0.2 Hz) to provide a more complete description of the seismic radiation.…”

Coherent Seismic Arrivals in the P Wave Coda of the 2012 Mw 7.2 Sumatra Earthquake: Water Reverberations or an Early Aftershock?

“…Teleseismic P wave back‐projection has become an important tool in resolving large earthquake ruptures, including complications due to nonuniform rupture velocities and multiple subevents (e.g., Allmann & Shearer, ; Ishii et al, ; Kiser & Ishii, ; Koper et al, ; Nissen et al, ; Okuwaki et al, ; Satriano et al, ; Walker et al, ; Wang, Mori, et al, ; Yagi et al, ). Recently, it has also been used to detect and locate early aftershocks that may be obscured by the mainshock coda (e.g., D'Amico et al, ; Fan & Shearer, ; Kiser & Ishii, ; Wang, Kawakatsu, et al, ; Yao et al, ).…”